NUWA Evidence record
Claude Opus 4.7
Cybench · Evidence record
Source snapshot, chart and review scope →
Original source charts and materials
Recorded measurements
- Cybench category pass@1
- 100%
2/2
Inspect Evals · Budget not specified
Assessment provenance
GLM-5.2 Risk Evaluation Report → · External report
Actual evaluation: SaferAI · Model record →
Method and scope
Task completion on 40 professional CTF problems.
- Evidence type
- Benchmark
- Assessment subject
- AI agent solving CTF tasks
- Primary dimension
- Capability
- Evidence snapshot
- 2026-10-29.1
Primary dimension follows the source’s measured construct. Metric-level interpretations may differ.
Source finding
The source reports 2/2. NUWA calculates the displayed 100% from those counts, rounded to two decimal places.
Limitations
Competition tasks and supplied guidance differ from operational attacks.
Small category sample; counts only; no estimated bar percentage.
Related research questions
Cybersecurity / Cyber task execution →
The source directly measures the stated construct under its protocol.
References
- [1] Andy K. Zhang, Neil Perry, et al.. Cybench: A Framework for Evaluating Cybersecurity Capabilities and Risks of Language Models. 2025-04-12. v4 [arXiv]; Author affiliations on the original paper title page
- [2] GLM-5.2 Risk Evaluation Report. 2026-08-02. v1.1 · revised 26 August 2026 ; p.6 Fig.2 printed bar label